High-performance Parallel Computing
Research Title: Development of cost-efficient parallel computing algorithms of governing equations related to turbulence.
​
Supervisor: Dr. Omer San, Oklahoma State University.
​
Goal: This is the side project I like to spend time on my own interest whenever I take a break from my primary research. The primary goal of this research is to enhance parallel computing performance of complex multiscale simulations, in terms of computation cost and acceleration.
​
Summary: In our codes, we use Message Passing Interface (MPI) domain decomposition and data exchange Application Programming Interface (API) and OpenMPI standard library. Our test cases are mostly in academic settings (since it's not a funded project, rather works on our own interest) which are 3D channel flow, Instabilities (Kelvin-Helmholtz, Rayleigh-Taylor, Ritchmyer-Meshokov) in 3D and 2D settings, 3D Taylor-Green vortex and so on. To learn MPI, interested readers are suggested the classical book "Using MPI: Portable Parallel Programming with the Message-Passing Interface" by Gropp, Lusk, and Skjellum (1999). Following a summary of our parallel computing research was presented as a poster in Coalition for advancing digital research and education (CADRE) conference, Stillwater, OK, USA.